Correlation of intra- and postoperative best active contact and its implication for STN-DBS outcome

Abstract

Objective

To assess the relationship between intraoperative test stimulation sites and postoperative best active contacts in STN-DBS for Parkinson's disease and to evaluate whether intraoperative VTAs predict long-term outcome.

Methods

We retrospectively analyzed 92 hemispheres in 46 patients undergoing bilateral STN-DBS. Coordinates of the best intraoperative stimulation point and the best active contact at one year postoperatively were compared using vector analysis. Corresponding VTAs were modeled and visualized in normalized MNI space using Guide XT and MATLAB-based tools. Response ratios were calculated based on motor improvement (UPDRS-III) at one year.

Results

The intraoperative and postoperative best stimulation points differed significantly (p < 0.001), with a consistent dorsolateral shift of the chronic stimulation site relative to the intraoperative test position. Group-level COG analysis showed a 1.61 mm Euclidean shift. VTA overlap between intraoperative and chronic stimulation sites was low (Dice index 0.26). Intraoperative VTAs did not predict clinical outcome. Mean motor improvement was 44 % (range 10–80 %). No significant difference in outcome was observed between patients with bilateral central trajectories and those with non-central implantations guided by intraoperative testing (p = 0.49).

Conclusion

Our findings demonstrate a systematic spatial difference between intraoperative and postoperative best active contact. Intraoperative VTAs, derived primarily from rigidity-based testing, did not predict global one-year motor improvement, underscoring the limited role of intraoperative testing as a universal predictor of long-term outcome. Imaging-based targeting achieved outcomes comparable to intraoperative test-guided implantation, supporting the feasibility of simplified, image-guided workflows in DBS planning.